Practical AI Application - The Thinking Lens

Sign in Subscribe

Practical AI Application

A collection of 9 posts

Beyond the Dashboard | Principle 11: Turn AI into a Judgment Multiplier

Data-Informed Decision-Making

Beyond the Dashboard | Principle 11: Turn AI into a Judgment Multiplier

AI is not your strategist; it multiplies your judgment. Automate discovery, keep humans in the decision loop, and treat judgment as the API. Clean hypotheses and consequence paths in, clarity out. Use AI to amplify decisions, not outsource them. Automate discovery, own decisions. Judgment is moat.

AI Doesn't Hallucinate. It Makes Mistakes.

Practical AI Application

AI Doesn't Hallucinate. It Makes Mistakes.

Calling AI errors “hallucinations” humanizes machines and inflates expectations. Language is the UI for trust; misuse becomes a shipped bug with churn, support cost, and legal risk. Treat wording like code: define terms, show process, and label errors precisely.

From “Don’t Do” to “Do Well”: Designing Instructions That Make AI Useful

Practical AI Application

From “Don’t Do” to “Do Well”: Designing Instructions That Make AI Useful

Instruction design turns a chaotic GPT into a reliable tool. Replace a 'don't do' list with a clear operating model: retrieve verbatim from the knowledge base, format consistently, handle exceptions, and test like software. Positive, explicit rules cut variance and improve UX.

Your Copy is Correct. It’s Also Forgettable.

Communication & Influence

Your Copy is Correct. It’s Also Forgettable.

Generative AI boosts efficiency but flattens voice. When tools push language toward the average, brands and leaders lose distinctiveness. Use AI for polish, not personality. Document tone, keep signature quirks, and build private style models to protect authenticity at scale. Keep your edge.

The Verde Archive bot

Practical AI Application

The Verde Archive bot

LinkedIn buries my old posts, so I built a lightweight AI lookup. Titles and links sit in JSON with ChatGPT-made tags. A custom GPT does prompt-based retrieval. Not full RAG, just fast recall. Next: automate updates via API. Guardrails keep it a helper, not a clone.

The “One Prompt” Fallacy: Why Voice, Not Prompts, Wins

Practical AI Application

The “One Prompt” Fallacy: Why Voice, Not Prompts, Wins

Stop chasing magic prompts. The fix is not a template. Define your problem, then train AI on your own writing to build a voice profile. Feed real samples, extract patterns, and revise in your style. Protect your data. Do not copy tone. Engineer sharper thinking, not generic outputs. Your voice wins.

LLMs, Longform, and Low-Code: What Worked (and Broke) in Real-World Tests

Practical AI Application

LLMs, Longform, and Low-Code: What Worked (and Broke) in Real-World Tests

LLMs aid with tight, scoped tasks: cut redundancy, multipass feedback, summary maps, brainstorming. But they struggle with long inputs, continuity and consistent voice. Useful for scaffolding and agents, not autopilot. Keep humans on structure, logic and style. Use models to spot patterns.

"Ahahaha! Busted! You used an em dash—must be ChatGPT!"

Communication & Influence

"Ahahaha! Busted! You used an em dash—must be ChatGPT!"

Ahahaha! Busted. You used an em dash, must be ChatGPT. I learned about hyphen, en dash, and em dash only recently. I asked ChatGPT, checked Treccani and Merriam Webster. I use AI a lot, but the voice is mine. Tools help me write clearer. The thoughts and the words are still mine. Fully human. True.

The Hidden Skill That Unlocks Generative AI

Communication & Influence

The Hidden Skill That Unlocks Generative AI

Generative AI’s power is tied to our ability to communicate with clarity; its output reflects the input. I saw this with my 8-year-old daughter: her vague prompt for a unicorn yielded generic results, but a detailed one created what she imagined. AI is a mirror.